Ординатура / Офтальмология / Английские материалы / Ocular Disease Mechanisms and Management_Levin, Albert_2010

Figure 42.2  Optic nerve injury induces intra-axonal and extra-axonal pathophysiology. An early event is wallerian degeneration, in which the distal axon degenerates. (Redrawn from Di Polo A. Mechansms of neural injury in glaucoma. In: Levin LA, Weinreb RN, Di Polo A (eds) Neuroprotection for Glaucoma: A Pocket Guide. New York: Ethis, 2007.)

There are likely other signals for RGC death besides neurotrophin deprivation. RGCs maintain viability for long periods of time when there is decreased axonal transport from compressive optic neuropathy or papilledema. Retrograde axonal transport is rapid, and the subacute time course by which RGCs die after axonal injury does not reflect the time course of interrupted retrograde axonal transport. RGC axotomy induces changes in responsiveness to neurotrophins independent of neurotrophin deprivation.72 Finally, removal of the RGC axonal target, and therefore, target-derived factors, causes very slow RGC death.57,73 These findings suggest that axotomy can signal changes at the cell body independent of neurotrophin deprivation. For example, an elevation in intracellular levels of the reactive oxygen species superoxide can occur independent of neurotrophin deprivation, and is necessary and sufficient for RGC death after axotomy.74

Wallerian (anterograde) and retrograde degeneration

Wallerian degeneration

Axonal injury in the CNS and PNS arising from traumatic, metabolic, toxic, inflammatory, and hereditary causes often results in a form of secondary axon pathology termed wallerian degeneration6 (see section on historical development, above). Focal injury to the axon leads to an orderly disassembly of the distal axonal stump in the hours to days following injury (Figure 42.2). At the cellular level, there is initial disassembly of the myelin sheath, followed by swelling of the axolemma, disorganization of neurofilaments and microtubules, and mitochondrial swelling. The remaining axonal fragments then undergo phagocytosis by glial cells and macrophages, followed by apoptosis of surrounding oligodendrocytes in the CNS.11

The directionality of wallerian degeneration was explored by Beirowski et al,75 who showed that, in mouse peripheral nerves, wallerian degeneration proceeds asynchronously (at different rates between different axons) and either anterogradely or retrogradely depending on whether it was a transection or crush injury that was incurred, respectively.75 Studies performed on dorsal root ganglion nerves had shown

Pathophysiology

•Local axonal injury

•Retrograde degeneration

•Wallerian degeneration

Figure 42.3  Axonal injury affects the retinal ganglion cell’s axon, dendrites, and soma, other neurons, and nonneuronal cells in the optic nerve and retina.

similar results, demonstrating that CNS transection injury causes anterograde degeneration. Wallerian degeneration may occur in a variety of diseases, including demyelinating diseases such as multiple sclerosis and Guillain–Barré, as well as following neurovascular insults, and neurodegenerative and infectious processes.12 As would be expected, wallerian degeneration is seen in optic nerve diseases.22,76–82

Dying-back degeneration

Another well-characterized type of axonal degeneration occurs by a process called dying back. As opposed to wallerian degeneration, which is thought to occur due to localized injury, dying-back degeneration occurs following a chronic and more generalized form of injury to the axon. It is a slow process, occurring over weeks to months, and involves degeneration of the axon from the synaptic end towards the cell soma in a retrograde fashion. It is known to occur in peripheral neuropathies, but in recent years evidence has accrued that it also is important for the pathophysiology of CNS degenerations (e.g., Alzheimer’s disease and Parkinson’s disease).7 Presumably a chronic optic nerve injury from a toxin, mitochondrial dysfunction, or nutritional deficiency could cause a parallel pathology to RGC axons.

Acute axonal degeneration

Though wallerian degeneration is the main form of axonal degeneration following axotomy, another degenerative process may occur in the acute stages of transection injury (beginning 20 minutes postinjury and lasting 5 minutes). Using in vivo time lapse imaging studies, Kerschensteiner and colleagues82 demonstrated that a process which they referred to as acute axonal degeneration (AAD) may occur prior to wallerian degeneration following dorsal root tran­ section in mice. AAD is thought to occur in the minutes following axotomy, causing axonal fragmentation in a bidirectional fashion in both the proximal and distal axonal stumps. This is thought to lead to retraction of the proximal axonal stump and subsequent wallerian degeneration of the distal end.82,83

Axonal degeneration is independent of soma degeneration

The mechanisms responsible for axonal degeneration are distinct from those causing apoptosis (Figure 42.3). Axonal degeneration often involves the calpains and/or the ubiquitin-proteosome system, unlike caspases in apoptosis.

Section 5  Neuro-ophthalmology Chapter 42  Optic nerve axonal injury

RGCs Brain

Figure 42.4  Axonal injury when apoptosis is blocked with bax knockout. The soma and proximal axon are spared (green), but the distal axon undergoes wallerian degeneration (red).

Figure 42.5  Axonal injury when wallerian degeneration is blocked in the WldS mutant. The distal axon is spared (green), but the proximal axon dies back and the soma undergoes apoptosis (red).

In recent years the distinction between apoptosis and axonal death has become clearer. It is now evident that, even though axonal death usually occurs following apoptosis (and vice versa), apoptosis is not a necessary requirement for axonal death. Axonal degeneration does not appear to occur due to a process of “starvation” from the cell soma, but rather seems capable of undergoing its own autonomous death process.

Blocking apoptosis

The DBA/2J Bax knockout mouse is just one of the experimental models which have helped to demonstrate this. DBA/2J mice develop elevated intraocular pressure and glaucoma as they age. DBA/2J mice crossed with mice with knockout alleles for Bax undergo axonal degeneration, but are protected from cell body apoptosis84 (Figure 42.4). These results demonstrate that axonal degeneration is distinct from apoptosis in glaucoma and that axonal death can occur in the absence of death of the cell soma.

Other clues have emerged from bcl-2 transgenic mice. Axotomy of nerves from mice containing the human bcl-2 transgene undergo axonal degeneration at a rate comparable to wild-type mice, but are protected from apoptosis of the cell soma. This demonstrates the importance of bcl-2 in preventing cell body death but its failure to protect from axonal degeneration.85 The pmn mouse, a mouse model of a motor neuropathy, is another example of the concept that axonal degeneration and death of the cell soma are distinct compartmentalized processes. Overexpression of the bcl-2 gene or inactivation of the bax gene in pmn mice prevents the death of the motor neuron cell bodies, but does nothing to prevent axonal degeneration.86 These mice continue to develop weakness and eventually die at a normal rate, despite protection from neuronal apoptosis.

Wallerian degeneration slow (WldS) mutants

Perhaps the most impressive evidence that axonal degeneration may occur independently from death of the cell soma was demonstrated using the wallerian degeneration slow (WldS) strain of mice. WldS mice have an autosomaldominant 85-kb tandem triplication mutation on chromosome 4 which confers delayed wallerian degeneration in both the PNS and CNS. In this strain of mice, axonal degeneration following injury or neurotrophin deprivation is delayed severalfold compared to wild-type mice, taking place several weeks after injury compared to only hours or days after. Remarkably, though axonal degeneration is delayed in these mice, death of the cell soma proceeds at a rate comparable to that of wild-type mice. That is, following injury, the process of apoptosis is not slowed or delayed in WldS mice.87,88 Axonal loss is slowed in glaucoma models in mice22 and rats89 when on a WldS background (Figure 42.5).

328

These results provide distinct and parallel evidence that axonal degeneration and apoptosis likely occur by very different and autonomous molecular mechanisms.

The genetic mutation responsible for the delayed Wallerian degeneration phenotype is an in-frame fusion of the N-terminal 70 amino acids of the E4 ubiquitin ligase Ube4b, an 18-amino-acid linker, and the full-length nicotinamide mononucleotide adenylyl transferase 1 (Nmnat1), involved with NAD+ synthesis.90,91 All parts are necessary for in vivo protection of axons from wallerian degeneration, and it is likely that direction of the fusion protein to a specific subcellular compartment is necessary for full effects.92–94 How this fusion protein blocks wallerian degeneration is an area of active research.93,95,96

Effects of axonal injury on other neurons

Axonal injury may also lead to the secondary degeneration of surrounding axons (which were uninjured by the primary insult). Partial optic nerve crush and transection injury models have demonstrated initial rapid injury of directly damaged axons, followed weeks or months later by degeneration of axons in adjacent areas.97–99 This was first demonstrated following traumatic injury to the CNS100 and is proposed to be a possible cause of continued RGC loss in optic neuropathies, particularly glaucoma, despite intraocular pressure control. Regardless of the type of injury, this secondary degeneration is believed to occur through a variety of mechanisms, including excitatory neurotrans­ mitter (glutamate) or oxygen free release by primarily injured axons, or by changes in extracellular ion concentrations, particularly potassium levels.51,97,101 Support for glutamate involvement stems from experiments using MK-801 (an N-methyl-d-aspartate (NMDA) receptor antagonist) following optic nerve crush injury, which has been shown to attenuate secondary degeneration.102

Finally, the impact of axonal degeneration is not only visible in close proximity to the initial lesion, but may also extend to target cells, located much further away. For example, the vast majority of RGCs establish terminal synapses in the lateral geniculate nucleus (LGN), and in several models of glaucoma, it has been shown that atrophy of cells in the LGN correlates with severity of RGC axon loss and of intraocular pressure (see Chapter 26).

Effects of axonal injury on nonneural cells

Though most work has focused on axonal injury and the subsequent effects on RGC death and/or survival, axonal disruption may also have consequences on surrounding retinal cells. Following optic nerve crush or spinal cord injury, for example, oligodendrocytes undergo apoptosis.

The loss of axonal contact and the decrease in neurotrophic factors are believed to lead to the initiation of the apoptotic program or of an atrophy-like resting state in oligodendrocytes following wallerian degeneration.103–106 Members of the tumor necrosis factor cytokine receptor superfamily and other molecules are involved in nonneuronal cell death after axonal injury.107,108

Other cells which are affected by axonal injury in the CNS include the resident microglia. These phagocytic cells increase in size and number and undergo activation in response to

Key references

wallerian degeneration, occurring several days later than the macrophage response in the PNS.11,109 Though they undergo activation following axonal injury,110 their phagocytic response nevertheless is limited and they incompletely clear myelin debris in the CNS.111 This is in contrast to PNS injury, where macrophages are actively recruited from the circulation and phagocytose myelin debris in an opsonindependent manner.112 Unlike in the PNS, there is a less profound influx of macrophages from the circulation during CNS injury.11

1.Levin L, Gordon L. Retinal ganglion cell disorders: types and treatments. Prog Retin Eye Res Sep 2002;21:465–484.

6.Waller AV. Experiments on the section of glossopharyngeal and hypoglossal nerves of the frog, and observations on the alterations produced thereby in the structure of their of their primitive fibers. Philos Trans R Soc Lond B Biol Sci 1850;140:423–429.

7.Raff MC, Whitmore AV, Finn JT. Axonal self-destruction and neurodegeneration. Science 2002;296:868–871.

8.Schwartz M, Yoles E, Levin LA. ‘Axogenic’ and ‘somagenic’ neurodegenerative diseases: definitions and therapeutic implications. Mol Med Today 1999;5:470–473.

11.Vargas ME, Barres BA. Why is Wallerian degeneration in the CNS so slow? Annu Rev Neurosci 2007;30:153–179.

12.Whitmore AV, Libby RT, John SW. Glaucoma: thinking in new ways–a role for autonomous axonal self-destruction

and other compartmentalised processes? Prog Retin Eye Res 2005;24:639–662.

24.Coleman M. Axon degeneration mechanisms: commonality amid diversity. Nat Rev Neurosci 2005;6: 889–898.

41.Quigley HA, Nickells RW, Kerrigan LA, et al. Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis. Invest Ophthalmol Vis Sci 1995;36:774–786.

57.Pearson HE, Thompson TP. Atrophy and degeneration of ganglion cells

in central retina following loss of postsynaptic target neurons in the dorsal lateral geniculate nucleus of the adult cat. Exp Neurol 1993;119:113– 119.

63.Quigley HA, Addicks EM. Chronic experimental glaucoma in primates. II. Effect of extended intraocular pressure elevation on optic nerve head and axonal transport. Invest Ophthalmol Vis Sci 1980;19:137–152.

74.Lieven CJ, Schlieve CR, Hoegger MJ, et al. Retinal ganglion cell axotomy induces an increase in intracellular superoxide anion. Invest Ophthalmol Vis Sci 2006;47:1477–1485.

84.Libby RT, Li Y, Savinova OV, et al. Susceptibility to neurodegeneration in a glaucoma is modified by Bax gene dosage. PLoS Genet 2005;1:e4.

87.Deckwerth TL, Johnson EM Jr. Neurites can remain viable after destruction of the neuronal soma by programmed cell death (apoptosis). Dev Biol 1994;165: 63–72.

88.Glass JD, Brushart TM, George EB, et al. Prolonged survival of transected nerve fibres in C57BL/Ola mice is an intrinsic characteristic of the axon. J Neurocytol 1993;22:311–321.

89.Beirowski B, Babetto E, Coleman MP, et al. The WldS gene delays axonal but not somatic degeneration in a rat glaucoma model. Eur J Neurosci 2008;28:1166–1179.

Clinical background

Historical development

Leber’s hereditary optic neuropathy (LHON) is a maternally inherited disease that presents with sudden or subacute nonsynchronous bilateral vision loss. Males in their second and third decade of life are typically affected. The classic visual field loss is a large and dense cecocentral scotoma usually associated with a decline of vision to greater than 20/200. LHON was first described by Von Graefe in 1858 and then characterized formally into a distinct clinical entity by Leber in 1871. Originally, it was believed to be X-linked and inherited with partial penetrance.1 Erickson in 1972 was the first to propose that LHON could have a maternal inheritance pattern from a mitochondrial mutation.2 Then in 1988, Wallace and colleagues confirmed the hypothesis by identifying a G to A mutation at nucleotide position 11778 in the mtDNA of nine pedigrees.3

Key symptoms and signs

LHON commonly manifests with acute or subacute painless central vision loss in one eye associated with dyschromatopsia. Within days, months, or rarely years, the second eye is similarly affected and the average interval time is 1.8 months.4 A few months after onset, the vision loss will typically plateau at or below 20/200. Acutely on clinical examination, the optic discs may appear hyperemic with a characteristic circumpapillary telangiectatic microangiopathy (Figure 43.1). The nerve fiber layer will be swollen without evidence of leakage of dye on fluorescein angiography, leading to the term pseudopapilledema. Over time, axonal loss of the papillomacular bundle (PMB) leads to temporal pallor and eventually severe and diffuse optic atrophy and large absolute cecocentral scotomas are found on visual field testing (Figure 43.2).

Genetics

Approximately 95% of the LHON cases of northern European descent are caused by the three most common mtDNA

Leber’s hereditary optic neuropathy

Alfredo Sadun and Alice Kim

mutations at nucleotide positions 3460, 11778, and 14484.5 These three mutations all result in an amino acid substitution in complex I of the respiratory chain. To date, more than 37 point mtDNA mutations have been identified.4 Of these point mutations, genes encoding for ND1 and ND6 appear to occur the most frequently. The ND1 and ND6 subunits are essential for mtDNA-encoded subunit assembly of complex I.4

However the presence of LHON mtDNA mutations does not necessarily correlate with vision loss. Only 50% of men and 10% of women who carry the LHON mtDNA mutation develop the optic neuropathy.6 However, over 80% of affected patients are male.7

Epidemiology

Studies in the north-east of England found that the minimum point prevalence of visual failure due to LHON was 3.22 per 100 000 and the minimum point prevalence for the LHON mtDNA mutation was 11.82 per 100 000. Therefore, LHON has a population prevalence similar to many autosomally inherited neurological disorders.8

Differential diagnosis

There are other conditions of inherited optic neuropathies, such as Kjer’s dominant optic atrophy (OPA1) and Wolfram syndrome (WFS1 gene on chromosome 4), that share similar clinical characteristics and hence must be differentiated from LHON. Though these inherited optic neuropathies are coded by somatic genes, it does not mean that they are not mitochondrial optic neuropathies. Indeed, we now know that their gene products do indeed interact in or with mitochondria. Metabolic optic neuropathies, including a large number of toxic and nutritional optic neuropathies, must also be considered in the differential diagnosis of LHON; nutritional deficiencies include insufficient levels of folic acid and vitamin B12 and toxic mitochondrial optic neuropathies such as ethambutol, chloramphenicol, and linezolid.9 Combinations of nutritional deficiencies and toxic exposure include tobacco–alcohol amblyopia and the Cuban epidemic of optic neuropathy.10–13

Treatment

As yet, no treatment for LHON has been proven effective. Antioxidants, such as vitamins C, E, and coenzyme Q10, have been offered to patients with LHON. This is based on theoretical grounds related to the electron transfer chain of oxidative phosphorylation, but without any demonstration of clinical efficacy. A coenzyme Q10 analog (idebenone) seems slightly more promising; it offers the additional advantage of transport into mitochondria and a few anecdotal case reports have demonstrated some clinical improvement.14,15

Prognosis

Visual recovery has been reported up to several years after vision loss and it is dependent on the age of onset and the specific mtDNA mutation. 14484 mtDNA mutations tend to have the best prognosis, whereas 11778 mtDNA mutations have the worst.16

Pathology

A few cases of molecularly characterized LHON have been studied histopathologically; however these tissues were

examined several decades after the clinical onset of disease.17 The most striking finding was the dramatic loss of retinal ganglion cells (RGC) and their axons, which constitute the nerve fiber layer and optic nerve (Figure 43.3). The centrally located small-caliber fibers of the PMB were completely lost, whereas the larger axons of the periphery were spared.18–20 Mitochondria tend to accumulate in the retinal nerve fiber layer (rNFL) and particularly just anterior to the lamina cribrosa.9 In the retrolaminar optic nerve, damaged mitochondrial accumulations occurred in demyelinated fibers, with activated astrocytes, glial cells, and lipofuscin-laden macrophages being observed near areas of relative axonal sparing.17,21 A wide variability in myelin thickness was seen along with evidence of some remyelination.17

As yet, no cases of histopathology of LHON have been obtained and examined during the acute phase of the disease. In this regard, rNFL analysis by optical coherence tomography (OCT) has been valuable and demonstrates significant thickening of the rNFL during the early and acute stages of LHON.22 These findings were most evident in the superior quadrant, followed by the nasal and inferior quadrants. Consistent with the PMB rNFL loss, less significant thickening was observed in the temporal quadrant. In the late stages, OCT revealed that the rNFL was significantly thinned and atrophic, with the temporal fibers being the most severely affected and the nasal fibers being relatively spared.23 Taken

as a whole, these studies suggest that the pathophysiology of LHON in the rNFL begins in the PMB and is associated with axonal swelling of the arcuate bundles. As atrophy sets in, the smaller-caliber fibers of the PMB are affected first. Later, the larger-caliber fibers of the arcuate bundles become thinned with comparative sparing of the nasal periphery.

Etiology

As yet, parts of the pathogenesis of LHON remain unclear. However, the underlying inherited basis of this disorder is understood. Though the primary etiologic cause is a mitochondrial genome (mtDNA) mutation, the presence of the LHON mtDNA mutation is necessary but not sufficient to lead to serious visual loss. As this is not a somatic mutation, the term carrier has been applied to those with the mutation but without significant visual loss. Many patients who are asymptomatic carriers may demonstrate subclinical disease, manifested as subtle dyschromatopsia.24 Affected patients are then said to have converted when they suffer abrupt and serious loss of vision. Thus, the penetrance of LHON is variable. Other genetic, epigenetic, and environmental factors appear to play a role in triggering the phenotypic expression of the disease. Studies of a large Brazilian pedigree of over 300 individuals have demonstrated this variable penetrance and the role of some environmental factors.10 In particular, smoking tobacco and drinking alcohol seem to increase

332

significantly the odds of conversion from carrier to affected status.25,26

Genetic factors

The three most common LHON mtDNA mutations are at nucleotide positions 3460, 14484, and 17788. Approximately 8–25% are due to the 3460 mutation, 10–15% account for the 14484, and 50–70% have the 17788 mutation.27 A higher amount of Asian LHON patients have the 11778 mtDNA mutation.28 In most LHON patients and family members, the mtDNA mutation is homoplasmic, containing only mitochondria with the pathogenic mutation. Approximately 14% of LHON patients carry both the mutant and wild-type DNA, a condition known as heteroplasmy. Studies have estimated that the heteroplasmy threshold for the phenotypic expression of LHON was 75– 80%.29 The prevalence of at least some heteroplasmy was 5.6%, 40%, and 36.4% for the 11778, 3460, and 14484 mtDNA mutations, respectively, in 167 unrelated LHON pedigrees.30

As stated earlier, approximately 50% of men and 10% of women who carry the LHON mtDNA mutation progress to develop the optic neuropathy.5 This male bias would suggest that the X chromosome plays a nuclear modulating role in the phenotypic expression of the disease. A previous study of 100 European pedigrees harboring all three mtDNA mutations identified a susceptibility locus on chromosome

Xp21-q21.31 An additive interaction was observed between this high-risk nuclear haplotype spanning markers DXS8090DXS1068 and mtDNA haplogroups. It found that 100% of individuals with both this chromosome X haplotype and a nonhaplogroup J background were visually impaired. Another novel susceptibility locus on chromosome Xq2527.2 has been identified in a large Brazilian pedigree that carry a homoplasmic 11778 mtDNA mutation on a haplogroup J background.32

Pathophysiology

Why are RGCs and their fibers in the optic nerve so vulnerable?

There are 1.2 million fibers of the optic nerve that arise from the RGCs of the retina.33 These RGC are highly concentrated in the perifoveal region and decrease in number moving out to the periphery. From each RGC is derived an

axon which merges into the overlying nerve fiber layer and eventually the optic nerve.34 Over 90% of the RGC are smaller parvo (P) cells which perform the discriminatory visual functions of high spatial frequency contrast sensitivity, spatial resolution, and color vision. The number of P cells is greatest in the macula. The other 5–10% of the RGC is comprised of the larger magno (M) cells that subserve low spatial frequency contrast sensitivity, depth perception, and motion. The PMB is composed of the smallest rNFL axons, and is dominated by P cells.35,36

The mitochondrial genome is very small, comprising only 16 569 basepairs, and most of the protein products necessary for mitochondrial biogenesis are not encoded in its genome. Hence, mitochondria replicate by organelle splitting and budding in the soma, near the nucleus and its chromosomes. Newly made mitochondria are transported down the axon to the terminals, pausing in this passage to provide adenosine triphosphate (ATP) at highly energy-dependent locations. Since the rNFL is unmyelinated and energyinefficient, mitochondria are plentiful in the prelaminar RGC axons, forming abundantly filled varicosities37 (Figure 43.4). The prelaminar and laminar regions of the optic nerve head represent another site of high energy dependency.38 Posterior to the lamina cribrosa, the axons are ensheathed by oligodendrocytes and become myelinated.34 Because of energy and temporally efficient saltatory conduction, the

334

number of mitochondria in the retrobulbar optic nerve is dramatically less, though they still congregate in the unmye­ linated gaps (nodes of Ranvier) and synaptic terminals. These unmyelinated areas require higher amounts of energy to restore the electrical potential. Thus, an inverse relationship exists between mitochondrial oxidative phosphorylation (OXPHOS) and myelination. Due to their high energy dependence, the unmyelinated rNFL and the prelaminar optic nerve are probably areas that remain the most vulnerable to mitochondrial dysfunction, whether in inherited conditions such as LHON or acquired metabolic disorders.

Mitochondria and oxidative phosphorylation

Mitochondria are double-walled organelles, found in all cells of the body except red blood cells. Located within the mitochondrial matrix are multiple copies of their own 16569-bp circular mtDNA.39 It encodes for 13 proteins, which are essential subunits of the OXPHOS complexes I, II, IV, and V. The rest of the approximately 80 subunit components are derived and transported from the nuclear genome. Since the OXPHOS chain is under both nuclear and mitochondrial genetic control, diseases that result from mitochondrial dysfunction can be inherited by both mendelian and maternal mitochondrial genetics40,41 (Figure 43.5).

ND1 ND6 ND4 CoQ Cyt C Inner

membrane

Pathophysiology

Figure 43.5  The mitochondrial chain of oxidative phosphorylation. The three most common pathogenic Leber’s hereditary optic neuropathy mutations 3460, 11778, and 14484, respectively affect the ND1, ND4, and ND6 subunits in complex I. Complexes I and II deliver electrons in parallel to complex III, after which serial electron transfer occurs down the chain to complex V.

Reactive oxygen species

OXPHOS is a vital part of metabolism. Byproducts of this highly efficient generation of ATP are reactive oxygen species (ROS) that are capable of damaging cellular enzymes, mtDNA, and membranes. Excess free electrons primarily spill from complexes I and III to react with molecular oxygen forming superoxide anion (O2–). The superoxide anion can undergo several transformations: conversion into hydrogen peroxide (H2O2) by manganese superoxide dismutase (MnSOD) or a reaction with nitric oxide (NO) to produce peroxynitrate (ONOO). Hydrogen peroxide can be further transformed by glutathione peroxidase (GPx) into water or react with transition metals to produce the hydroxyl radical (OH) via the Fenton reaction.

The excess ROS may go on to injure the Fe-S centers of complexes I, II, and III and proteins of the tricarboxylic acid cycle enzymes.42,43 Furthermore, ONOO can damage complex I through thiol-nitrosylation and via the addition of nitrate to tyrosine residues of complex I and MnSOD.44,45 The ROS can damage the mtDNA itself, resulting in multiple gene deletions, and instigate lipid peroxidation of the mitochondrial membranes. It is likely that this toxic ROS-rich environment provided selection pressure eons ago, such that much of the original mtDNA derived from primitive prokaryotes “moved” to the nuclear chromosomes. For reasons having to do with transmitochondrial membrane transport, only mtDNA encoding for 13 protein products remained in the mitochondria still vulnerable to the damaging effect of the ROS.

release.47,48 Cyt c, which functions as an electron carrier to complex IV, can then pass through the mitochondrial membrane into the cytosol to form a complex called “apoptosome.” This cleaves procaspase 9 to become caspase 9, setting off a chain of caspase reactions that leads to the destruction of somatic DNA and apoptosis.46

Biochemical effects of the mtDNA mutations

Complex I of the respiratory chain is composed of 39 nuclear and seven mitochondrial encoded subunits.49 The three most common pathogenic LHON mutations, 3464, 11778, and 14484, respectively affect the ND1, ND4, and ND6 subunits in complex I. Cellular models using transmitochondrial hybrids called cybrids have been widely used in order to study the effect of the mtDNA mutations on complex I activity. A significant reduction in complex I activity was observed with the 3460/ND1 mutation, whereas minimal reductions were noted with the 11778/ND4 and 14484/ ND6 mutations.50 Rotenone is a powerful complex I inhibitor that acts as a coenzyme Q intermediate antagonist and the mitochondrial activity of each LHON mtDNA mutations is variously affected by it. The 11778/ND4 and 14484/ND6 mutations seem to have decreased sensitivity to rotenone. The biochemical effect of other product inhibitors such as quinol and myxothiazol have also been evaluated and all three common mutations are observed to have increased sensitivity.50 This suggests that LHON mutations impair the interaction of complex I with the Q substrate.

Mitochondrial role in apoptosis

There are many parallel pathways that converge and diverge in the regulation of the execution of apoptosis. However, one important process appears to use mitochondria as a lynch pin that determines cellular death by apoptosis.46 OXPHOS, ROS, the mitochondrial inner membrane potential, and calcium fluxes are all involved in the regulation of mitochondrial permeability. If circumstances allow a critical threshold to be reached, small gates called mitochondrial permeability transition pores can open, allowing the release of proapoptotic factors. These gates are also under nuclear control by proapoptotic signals such as Bax, Bak, Bim, and Bid and the anti-apoptotic protein Bcl-2, which blocks cyt c

Proposed LHON pathogenesis that leads to RGC death

There are two main consequences of complex I dysfunction as evidenced in LHON: impaired efficiency of the respiratory chain manifested as a reduction in ATP synthesis and increased ROS production. In turn, the decreased ATP production may lead to slowed axoplasmic transport, particularly of mitochondria. Consequently, axonal swelling and stasis develop in the prelaminar unmyelinated portion at the optic nerve head. The mitochondria, which must make their way from the RGC soma down the long axons of the optic nerve to their synaptic terminals in the brain, have a short lifespan, in the order of a week or two. Thus, their failure to arrive in a timely fashion would ultimately lead to retrograde

Box 43.1  Key points

•The majority of Leber’s cases are caused by three common mtDNA point mutations at nucleotide positions 3460, 11778, and 14484, which respectively affect the ND1, ND4, and ND6 subunits in complex I of the respiratory chain

•Leber’s hereditary optic neuropathy has a variable penetrance. Other genetic, epigenetic, and environmental factors appear to play a role in triggering the phenotypic expression of the disease

•All mitochondrial optic neuropathies similarly manifest with cecocentral scotomas, dyschromatopsia, poor vision, and a loss of high spatial frequency contrast sensitivity. Leber’s hereditary optic neuropathy is perhaps the best-studied model of mitochondrial deficiency

wallerian degeneration of the optic nerve fibers. Compounding this loss of energy and axonal stasis, the impairment in complex I activity results in the accumulation of ROS which can, at a certain threshold, lead to RGC and axon death by apoptosis. The small fibers of the PMB are uniquely vulnerable to this process, responding with an early and rapid wave of cellular death.

These described mechanisms are likely to be critical in the different inborn and acquired impairments of mitochondria

and result in the progressive decompensation of the RGC and optic nerve system. Hence, it is not so surprising that all mitochondrial optic neuropathies can manifest similarly with cecocentral scotomas, dyschromatopsia, poor visual acuity, and a loss of high spatial frequency contrast sensitivity. LHON is perhaps the best-studied model of mitochondrial deficiency and recent discoveries, such as the demonstration of subclinical disease, are likely to have applications in other related mitochondrial optic neuropathies of inherited (Kjer’s), metabolic, and toxic etiologies.

Since the discovery of point mtDNA mutations in LHON, progress has been made in elucidating the pathogenesis of the disease (Box 43.1). However, much has yet to be clarified about the male predilection, incomplete penetrance, and abrupt age-related onset of vision loss. Cellular and animal models will likely pave the way for understanding the complex biochemical interactions, nuclear and mtDNA genomic factors, and pathologic progression of the disease. Future therapeutic modalities for LHON could be directed toward a genetic approach to prevent optic atrophy, the discovery of ROS scavengers, and antiapoptotic strategies once the onset of vision loss transpires. However since many of these therapies will require the resolution of complex biochemical and biological issues, currently we look forward to the development of serum biomarkers and psychophysical measures to identify patients with early disease or at least subclinical evidence that decompensation may be near.

3.Wallace DC, Singh G, Lott MT, et al. Mitochondrial DNA mutation associated with Leber’s hereditary optic neuropathy. Science 1988;242:1427–1430.

4.Carelli V, Ross-Cisneros FN, Sadun AA. Mitochondrial dysfunction as a cause of optic neuropathies. Prog Retin Eye Res 2004;23:53–89.

5.Mackey DA, Oostra R-J, Rosenberg T, et al. Primary pathogenic mtDNA mutations in multigeneration pedigrees with Leber hereditary optic neuropathy. Am J Hum Genet 1996;59:481–485.

7.Newman NJ, Lott MT, Wallace DC. The clinical characteristics of pedigrees of Leber’s hereditary optic neuropathy with the 11778 mutation. Am J Ophthalmol 1991;111:750–762.

9.Sadun AA, Carelli V, Salomao SR, et al. A very large Brazilian pedigree with 117 788 Leber’s hereditary optic neuropathy. Trans Am Ophthalmol Soc

2002;100:169–178; discussion 178–179.

17.Carelli V. Leber’s hereditary optic neuropathy. In: Schapira AHV, DiMauro

S (eds) Mitochondrial Disorders in Neurology, 2nd edn. Boston: Butterworth-Heinemann, 2002:115– 142.

18.Sadun AA, Kashima Y, Wurdeman AE, et al. Morphological findings in the visual system in a case of Leber’s hereditary optic neuropathy. Clin Neurosci 1994;2:165–172.

19.Sadun AA. Acquired mitochondrial impairment as a cause of optic nerve disease. Trans Am Soc 1998;XCVI:881– 923.

20.Sadun AA, Win PH, Ross-Cisneros FN, et al. Leber’s hereditary optic neuropathy differentially affects smaller axons in the optic nerve. Trans Am Ophthalmol Soc 2000;98:223–235.

22.Barboni P, Savini G, Valentino ML, et al. Leber’s hereditary optic neuropathy with childhood onset. Invest Ophthalmol Vis Sci 2006;47:5303–5309.

23.Barboni P, Savini G, Valentino ML, et al. Retinal nerve fiber layer evaluation by optical coherence tomography in

Leber’s hereditary optic neuropathy. Ophthalmology 2005;112:120–126.

24.Ventura DF, Quiros P, Carelli V, et al. Chromatic and luminance contrast sensitivities in asymptomatic carriers from a large Brazilian pedigree of 11778 Leber hereditary optic neuropathy. Invest Ophthalmol Vis Sci 2005;46: 4809–4814.

25.Sadun AA, Carelli V, Salomao SR, et al. Extensive investigation of large Brazilian pedigree of Italian ancestry (SOA-BR) with 117788/haplogroup J Leber’s hereditary optic neuropathy (LHON). Am J Ophthalmol 2003;136:231–238.

26.Sadun F, De Negri A, Carelli V, et al. Ophthalmologic findings in large pedigree of 11778/haplogroup J Leber’s hereditary optic neuropathy. Am J Ophthalmol 2004;137:271–277.

46.Sadun AA, Carelli V. The role of mitochondria in health, aging, and diseases affecting vision. Br J Ophthalmol 2006;90:809–810.